Main site

Media & Entertainment AI agent · Residuals

Residuals & Participations AI Agent

Assemble the residual calculation and re-perform the statement against the agreement it is owed under, track the assumption chain and the clocks that run, and hold both for the officer who signs.

Typical delivery
4–6 weeks
Deployment
Your stack
Officer signs
Pre-issuance
After launch
Agent Care

What this agent does

Assembles the calculation, not the statement

In

  1. 01

    A title is licensed into a new market, and the use is read from the distribution systems on file.

  2. 02

    A use is mapped to the market, the formula and the base the guild agreement in force applies to it.

Reason

  1. 03

    A window opens, and the residual is recomputed and reconciled to what was reported and what was paid.

  2. 04

    A title changes distributor, and the chain is checked for an executed assumption agreement.

  3. 05

    A payment carries the agreement version, the market and the run it was calculated under, wherever it goes.

Decide

  1. 06

    A due date that differs by market and by guild is calendared separately and driven to the earliest one.

  2. 07

    A statement is re-performed from the contract's own definitions and held for the participations officer.

Out

  1. 08

    A finding is retained with its class — mechanical, interpretation or equity — and the evidence under it.

  2. 09

    A write happens only inside the approval boundaries agreed during implementation, never ahead of approval.

Product statement

The agent assembles and reconciles; a participations officer issues and owns the statement, and a guild or participation auditor tests it.

Example workflow

One title, use to statement

  1. 1Use receivedDistribution report, licence record, viewership file or payment record
  2. 2Agreement assembledThe guild agreement in force, the market, the formula and the base, each with its version
  3. 3Residual proposedMarket, formula, base, amount due and confidence
  4. 4Controls appliedAssumption-chain checks, due-date tests, agreement-version checks and confidence threshold

No human action required

Stages 1 to 4 run unaided, and nothing is issued at any of them — the agent is assembling, and the officer's lane opens at the confidence gate.

5DecisionBranches at the confidence threshold
High confidence

Goes to the participations officer.

Low confidence

Adds a residuals-manager read first.

Officer approval

The calculation is held with its agreement version, its evidence and the confidence.

Approve · Amend · Send to residuals review

Approved — into the statement

  1. 6Payment systems updatedOnly where write access and approval policy allow it
  2. 7Outcome evaluatedUnderpayments found on audit, late-payment damages and claims raised after issue

Amendments

An underpayment found after the clock has run is unrecoverable, and it is still counted.

What should not run autonomously

Human approval stays in control

Outside the boundary — human approval required

8 items

  • Issuing or signing the participation statement.
    • Certifying that a residual has been paid in full.
    • Agreeing that a transferee has assumed an obligation.
    • Opining on what a profit definition's terms mean.

  • Automation boundary

    Agent acts unaided

    • ✓Recompute the residual under the agreement in force.
    • ✓Reconcile what was reported against what was paid.
    • ✓Track the chain and flag a transfer with no assumption on it for the named owner.
    • ✓Run the clocks, raise the gaps and hold the file for the officer.

    Any write happens inside the boundaries agreed at implementation, never ahead of approval.

    • Executing an assumption or a security agreement.
    • Demanding an audit or agreeing a tolling extension.
    • Settling or waiving a participation claim.
  • Changes to calculation, chain or approval rules.

Example output

One residual, annotated

Everything the agent proposes is attached to the record it was calculated from.

Residual output · single use

Illustrative example
Title
Drama series, season one
Use
Domestic reuse on the initial-release service
Market
High-budget SVOD
Agreement cited
2026 SAG-AFTRA agreement
Confidence
93%
Determination
Due — assumption chain complete

As received

Taken from the distribution report and the agreement on file — nothing on this side is written by the agent.

Evidence used

  • Executed assumption
  • Distribution report
  • Agreement version

Why this determination

The transferee signed and the guild approved it — the call the officer approves.

What the score decides

Below the configured threshold the calculation picks up a residuals-manager read.

Action

ApproveAmendSend to residuals review

Value

Where AI adds value

The same four claims, placed at the point in the workflow where each one applies.

Where the value lands

Value 01 – 04

Every use

From the distribution report

03Calculation

Apply the agreement in force

Read the agreement the use falls under: the 2026 WGA MBA runs to 1 May 2030, while IATSE's 2024 Basic Agreement runs to 31 July 2027.

01Approved path

Assuming does not release

A distribution assumption agreement adds a payer; SAG-AFTRA's television form says nothing in it releases the producer.

02Human review

Read every guild separately

SAG-AFTRA's 2026 agreement, effective 1 July 2026, lifts its top foreign subscriber tier; the WGA's 2026 MBA moved foreign streaming another way.

04

Build an evidence trail

The title, the agreement it was calculated under and the officer who issued stay on the statement.

Integrations

Typical integrations

Five system groups connect to the same agent. Which of them are in scope is decided in discovery.

  • Distribution and sales

    • Rights and licence systems
    • Distribution reporting

  • Residuals and payroll

    • Entertainment Partners
    • Cast & Crew
    • GreenSlate

  • Participations

    • Statement and ledger systems
    • Studio finance systems

Agent

Residuals & participations

  • Reads the use
  • Recomputes the residual
  • Holds for the officer

  • Documents

    • Assumption agreements
    • Contracts
    • profit definitions

  • Observability & evaluation

    • OpenTelemetry
    • Langfuse
    • Supported monitoring/evaluation sources

Integration availability depends on the client's existing systems and API access.

Agent controls

Six layers between the model and the statement

Six checks, narrowing as they go. What comes through the last one is named in the map below.

L6 · Outermost — last line of defence

Inward → L1 · closest to the model

L6Rollback / safe modeRevert to data assembly when evaluation or production signals degrade.Roll back
L5Version monitoringTrack model, prompt, agreement-version and formula-table changes.Track
L4TraceabilityRecord the use, the agreement cited, the calculation and the approval time.Record
L3Officer approvalHold the calculation for the named officer; a statement already issued is past every layer here.Gate
L2Policy guardrailsTest each calculation against the agreement's own market and formula tables; a failure returns it unreconciled.Restrict
L1Confidence thresholdsRoute low-confidence calculations to a residuals-manager read first.Require review
Model coreResidual proposed — market, formula, base, amount due and confidence
L1 – L2
Test whether a calculation may stand
L3
Puts the statement in an officer's hands
L4 – L5
Keep the title and the agreement behind it
L6
Reverts to data assembly when signals degrade

How Nestack evaluates it

Evaluate the calculation workflow — not only the finished statement.

Coverage runs the whole depth of the workflow, and every layer is cut by slice.

Surface — the statement the participant reads

  1. E1Final-output evaluationDid the amount follow the market, the formula and the base cited?
  2. E2Step-level evaluationDid the agent use the right agreement, version and market?
  3. E3Tool evaluationDid it read the correct title and the correct distribution report?
  4. E4Confidence calibrationDo low-confidence calculations actually attract more amendments?
  5. E5Slice evaluationHow does performance change across specific markets and guilds?
  6. E6Business outcomeHow many payments were found short on audit, or amended after issue?

Floor — the claim the participant loses to the clock

Failure modes

Where each failure originates in the agent

Seven failure modes, placed at the stage each one originates.

Agent lifecycle

Direction of processing →

  1. 01 · Retrieval

    1 mode

    • ZQ-03

      Unreported licence revenue

      A sublicence never reaches the distribution report.

    Stage gathers

    Distribution reports, agreements and payment records

  2. 02 · Reasoning

    2 modes

    • ZQ-04

      Superseded agreement used

      A retired cycle's factor is read onto a 2026 reuse.

    • ZQ-06

      Cross-guild substitution

      One guild's mechanism is applied to another's.

    Stage proposes

    Market, formula, base, amount due and confidence

  3. 03 · Tool / write

    2 modes

    • ZQ-02

      Assembly taken as issuance

      A reconciliation is treated as a statement.

    • ZQ-05

      Incontestability clock missed

      A window is run from receipt, not from issue.

    Stage writes

    Only where write access and approval policy allow it

  4. 04 · Output

    1 mode

    • ZQ-01

      Underpaid residual

      A payment goes out short against the agreement.

    Stage returns

    The statement the participant is asked to accept

  5. 05 · Change / Version

    1 mode

    • ZQ-07

      Silent formula drift

      A model or table change alters what is computed.

    Stage tracks

    Model, prompt, agreement versions and formula tables

  • Sev-1 · a statement issues past the boundary
  • Sev-2 · a short payment reaches the guild
  • Sev-3 · source degrades, use routes to review
See the complete Media & Entertainment failure-mode catalogue

Affected slices

Accuracy can hide one exposed cohort

A programme-wide accuracy figure across the guilds can look acceptable while a small number of title cohorts carry most of the underpayments. Nestack reports the underpayment rate by slice, not only in total.

Slice performance — reported separately, not only in aggregate

Illustrative example
SliceFailure rateLiftLift vs. thresholdStatus
Transferred library titles7.8%3.8×Review
Foreign streaming reuse5.4%2.6×Review
Basic cable reruns3.6%1.8×Watch
Current-cycle first runs1.5%0.7×Normal

Bar: underpayment-rate lift vs. first-run baseline · scale 0–4.0× · tick marks the 2.0× review threshold

2 of 4 slices over threshold

Evidence-linked improvement

A cycle closes on a test, not a memo

The loop shuts when the missed payment is a regression case. That is what the next statement issued is measured against.

Improvement cycle · five stages

Switchback — the path turns at Improve and returns at Learn

  1. 01

    Detect

    Underpayment rate rises in a title slice.

  2. 02

    Diagnose

    The picture that changed distributor and lost its payer is traced back to the transfer nobody papered.

  3. 03

    Improve

    Any change goes out numbered, with the titles that caused it attached.

  4. 04

    Verify

    A statement case still failing is enough to hold the release back.

  5. 05

    Learn

    One case added to the suite, one line added to the assumption record.

  6. Learn → Detect

    The return edge. The next detection runs against a suite one underpayment case longer.

Typical build scope

Twelve workstreams across six weeks

The build scope read against the delivery timeline. Week structure follows the six-week plan — discovery, sources, calculation workflow, evaluation, integration, then production validation and handover.

Workstream

Week 1Week 2Week 3Week 4Week 5Week 6

  1. 01Residuals workflow discovery and boundary definition.Week 1
  2. 02Distribution and payment source review.Week 1, Week 2
  3. 03Agreement, market and formula-table rule mapping.Week 1, Week 2
  4. 04Use ingestion and market classification.Week 2
  5. 05Calculation logic and agreement binding.Week 2, Week 3
  6. 06Confidence scoring and gap routing.Week 3
  7. 07Officer approval workflow.Week 3, Week 4
  8. 08Distribution and payment integration.Week 5
  9. 09Calculation and chain cases.Week 4, Week 5
  10. 10Guardrails and issuance controls.Week 4
  11. 11Title-trail instrumentation.Week 5, Week 6
  12. 12Deployment, documentation and Agent Care handover.Week 6

12 workstreams · 6 weeks · bar shows the weeks a workstream is active — several run in parallel

Final scope and sequence confirmed in discovery

Engagement tiers

What each tier includes

Rows are the capabilities named in each tier's scope. Higher tiers include everything below them.

Capability✓ in scope · — not at this tierPilotOne title, one guildProductionProduction residual systemsAdvancedMultiple slates / guilds
Introduced at Pilot
Calculation to your agreements✓✓✓
Officer approval✓✓✓
Calculation-accuracy baseline✓✓✓
Introduced at Production
Reporting by guild—✓✓
Approval workflow in your systems—✓✓
Approved write-back—✓✓
Distribution-data integration—✓✓
Introduced at Advanced
Multi-guild agreement sets——✓
Multi-stage studio approvals——✓
High title volume——✓
Multi-agreement residual controls——✓
Build priceFrom $5,000From $8,000Custom quote
Final build price
Confirmed after discovery based on integrations, workflow complexity, transaction volume, approval controls and deployment requirements.
Separate from build
Build pricing is separate from recurring Agent Care, which covers managed monitoring, evaluations, incidents and verified improvements after launch.

What we need from you

What you bring, and what we build with it

Each input maps to a piece of build scope and a week in the delivery timeline.

  1. 01Your titles and their agreement versions

    Use ingestion and market classification

    Week 1

  2. 02Representative statements already issued

    Calculation baseline and reconciliation checks

    Week 2

  3. 03Your assumption agreements and chain of title

    Agreement, market and formula-table rule mapping

    Week 1

  4. 04Access to relevant APIs, feeds or exports

    Distribution and payment assessment, then integration setup

    Week 2

  5. 05Statements you would not want audited

    Underpayment cases and failure-mode testing

    Week 4

  6. 06What no statement may omit

    Confidence scoring, gap routing, guardrails and issuance controls

    Week 3

  7. 07Named participations officers to approve

    Officer approval workflow, then pilot and production validation

    Weeks 5–6

Nothing else is required

Deployment, documentation and Agent Care handover are ours.

Delivery timeline

Four phases across six weeks

Each band spans the weeks its phase truly takes, which is why the fifth carries two of them at once.

Phase

W1

W2

W3

W4

W5

W6

  1. DiscoveryW1
  2. BuildW2 – W3
  3. EvaluateW4 – W5
  4. Pilot & LaunchW5 – W6

Week focus

W1Residuals workflow discovery, agreement mapping and the automation boundary

W2Distribution and payment integration and the calculation baseline

W3Calculation workflow, confidence logic and issuance controls

W4Evaluation suite, assumption-chain checks and failure-mode testing

W5Payment-system integration, pilot titles and targeted corrections

W6One reporting cycle run under the participations officer, then handover

Reading the band
Each bar covers only the weeks its work is named in. The week 5 overlap is real work, not padding.
At the end of W6
Once the cycle validates, Agent Care assumes the running agent.
Duration
Six-week plan shown · typical delivery 4–6 weeks depending on scope confirmed in discovery.

Next step · Media & Entertainment AI agent

Build a residuals agent around the agreement each payment is owed under.

A statement that has issued starts a clock that runs whether or not it was right. Show us your titles, your agreements and who issues — the officer signs, and a guild auditor tests it.

Typical delivery
4–6 weeks
Deployment
Your distribution data
Officer signs
Pre-issuance
After launch
Agent Care available

Nestack Agents · Residuals & participations

AGT-ME-15 · Agent Care available after launch

Build a residuals agent

Four questions so the call starts from your actual setup — then pick a time that suits you.

About you
Your titles

No obligation. We’ll map the workflow, name the automation boundary and tell you if an agent is the wrong answer.